The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 59:1081-1088 (2004)
© 2004 The Gerontological Society of America
Age-Related Changes in Metabolic Parameters of Nonhuman Primates
Xenia T. Tigno,
Georgielle Gerzanich and
Barbara C. Hansen
Obesity and Diabetes Research Center, Department of Physiology, School of Medicine, University of Maryland at Baltimore.
Address correspondence to Xenia T. Tigno, PhD, Department of Physiology, Obesity and Diabetes Research Center, School of Medicine, University of Maryland at Baltimore, 10 South Pine St., MSTF Rm. 6-00, Baltimore, MD 21201-1192. E-mail: xtigno{at}yahoo.com
 |
Abstract
|
|---|
Some physiological measures change with age, but the existence of age-related disorders such as type 2 diabetes raises questions about which patterns reflect progressive pathology and which are manifestations of aging. Here we report a retrospective investigation of age-related physiological changes in rhesus monkeys that developed diabetes (D group, n = 65) or exhibited healthy aging (N group, n = 88). Data were available on clinical chemistries, hematology, glucose tolerance, and insulin sensitivity based on oral and intravenous glucose tolerance tests and euglycemic, hyperinsulinemic clamp assessments. Individuals contributed data for an average of 7.6 years, when they were between 5 and 30 years of age. Only glucose disappearance rate, insulin sensitivity rate, and high density lipoprotein levels changed significantly with age in the nondiabetic group. In the diabetic group, significant decreases in glucose tolerance were evident by middle age (age 14 y), and fasting insulin first increased before diabetes was diagnosed, and then declined with advancing age.
DEFINING the boundaries between "normal biological aging" and aging-associated disease has always been difficult. With the decline in physiological functions as well as the increase in prevalence of degenerative diseases with advancing age, it is often debated whether aging and disease constitute a dichotomy or merely a continuum. The same may be argued in the case of aging and metabolic disease. It is often presupposed that glucose tolerance deteriorates as one ages. Based on parameters such as fasting plasma glucose (FPG), glucose (G), and insulin (I) levels during an oral glucose tolerance test (OGTT), and glucose uptake during euglycemic clamps (M-rate), studies comparing purportedly healthy elderly participants with younger participants have demonstrated an elevation of OGTT-G and reduction of M-rates, albeit modest, among elderly persons, as well as a progressive increase in fasting plasma glucose with progression of aging (1–7).
Data involving human subjects are, however, fraught with difficulties, such as the presence of multiple confounders that may directly or indirectly impact upon the variables of interest. These include, among others, lifestyle, dietary composition, degree of adiposity, physical activity, alcohol consumption, substance abuse, and ethnicity. While variability due to these factors and their interactions may be adjusted for in the analysis of the data, a laboratory- controlled environment provides a milieu where most of these factors are absent or tightly regulated. Furthermore, classification of human subjects as being "normal" or "diabetic" also depends on prevailing criteria, such as those recommended by the World Health Organization (WHO) or the American Diabetes Association (ADA). A review by Elahi and colleagues (7) summarized the results from 25 prevalence studies where both fasting glucose and glucose tolerance tests were performed as advocated by the WHO. He calculated that if only FPG were considered as the criterion, as suggested by the ADA, "a very high percentage (ranging from 11%–80%) of the diagnosable diabetics are missed if the glucose tolerance test is not considered." This suggests that previous studies using only FPG as the criterion may have missed a number of diabetics (who would have been diagnosed based on OGTT) and probably included these subjects among their normal population. Because of these difficulties in classification, Rowe and Kahn have divided the normal elderly group into those who are "usual," which consists of a more heterogeneous group of individuals who are nondiseased and show the dominant trends in the general population, and those who have aged "successfully," meaning those who demonstrated no loss of function or impairment with regard to the variable under study (8). The present study attempts to obviate these difficulties by analyzing the data obtained from a primate facility at the University of Maryland, where complete medical records dating back over more than 20 years are available. Many of the monkeys spontaneously develop diabetes and all its symptoms, akin to type 2 diabetes in humans. The natural history of the development of obesity and diabetes in this colony has been previously described (9). In this controlled environment, variations due to lifestyle, diet, physical activity, illness, medication, and other extraneous stressors are effectively removed. Furthermore, because of periodic metabolic assessment and close monitoring of the monkeys (twice per year), and the luxury of retrospect, misclassification of subjects as to their being "cases" or "control" is highly unlikely. Monkeys that are predisposed to developing diabetes, meaning those that manifest the characteristics of the metabolic syndrome, can easily be identified. Thus, in this study, in contrast to previous analyses performed by this and other laboratories, we have only included among our nondiabetic subjects, monkeys that have or are aging "successfully" as far as absence of diabetes is concerned. In contrast, most past studies have used for their comparator group the "usual" elderly person. In this study, not all subjects with fasting glucose levels below 126 mg/dl were included among the normal group if they were considered to exhibit the metabolic syndrome. Because past studies have for their control subjects a more heterogeneous group, the "usual" metabolic changes believed to be intrinsic to aging may have been due to the contribution of subjects in the preclinical phase of the disease. For this reason, previous results may be very different from the present study. In addition, since we could identify the monkeys that eventually developed diabetes, the data from these animals before conversion did not contribute to the "normal" data set, but was analyzed instead as part of the data of the diabetic group at that particular age. Since our normal group contained neither animals with the metabolic syndrome, nor animals that at some point eventually became diabetic, we hope that we have identified physiologic changes in "successful" aging, distinct from changes due to disease.
|
METHODS
|
|---|
One hundred fifty-three (153) monkeys, age ranging from 4.78 to 29.48 years, were studied longitudinally over a period of approximately 20 years. Body weights ranged from 4.0 to 29.7 kilograms. The animals were housed individually and maintained under a 12-hour light and dark cycle, with an ambient temperature of 22°C and were allowed ad libitum access to food and water. The diet consisted of standard laboratory primate chow (LabDiet 5038; Purina Mills, St. Louis, MO), which provided 
18% of calories as protein, 13% as fat, and 69% of calories as carbohydrates. The monkeys were classified as belonging to either the Diabetic (D) group or the Nondiabetic (N) group according to the following criteria: Diabetic subjects were those monkeys that during the period of the study attained at least two values of FPG 
126 mg/dl, which is based on both the ADA (10) and WHO (11) guidelines. Nondiabetic subjects consisted of adult monkeys studied longitudinally that did not develop diabetes during their lifetime or during the period of the study, nor develop the metabolic syndrome, which is defined for monkeys as having four of five of the following characteristics: a fasting plasma glucose of greater than or equal to 80 mg/dl, body fat greater than or equal to 25%, fasting triglyceride (TG) levels greater than or equal to 80 mg/dl, blood pressure levels above 120/75 mmHg, and fasting high density lipoprotein (HDL) cholesterol values of below 60 mg/dl. Monkeys with fewer than four of the above criteria were not excluded from the nondiabetic category.
Semiannual measurements of key variables were obtained to assess the metabolic status of each animal. Plasma samples were obtained after a 16-hour fast, with ketamine hydrochloride (10 mg/kg body weight) as anesthetic. Fasting plasma glucose levels were obtained using the glucose oxidase method, while fasting plasma insulin (immunoreactive insulin, IRI) values were determined using a radioimmunoassay procedure performed at LINCO Diagnostic Services (St. Charles, MO). Hematologic and blood chemistry assays were performed by a commercial laboratory, and TG and lipoprotein fractions were determined by a different laboratory. Glycated hemoglobin (HbA1C) was obtained using a Bayer DCA 2000+ analyzer (Fisher Scientific, Pittsburgh, PA) and reagents. Intravenous glucose tolerance testing was performed as described in a previous article (12) and consisted of administering a bolus of 0.5 ml/kg body weight dextrose 50% to give a final dose of 0.25 g glucose/kg body weight. Insulin and glucose levels were measured at baseline, then at 1, 2, 3, 5, 7, 10, 15, 20, 30, 40, 50, and 60 minutes after glucose administration. KG (glucose disappearance rate) was calculated using the following formula:
|
where KG represents the log-linear decline of glucose levels. The acute insulin response was obtained by averaging the insulin values over the first 10 minutes after glucose infusion. Lean body mass was estimated using the tritiated water dilution method as previously described (13). Plasma samples were collected 60 minutes following intravenous administration of 4 µCi/kg body weight of tritiated water at which time the dose was assumed to have achieved equilibrium and total body water could be estimated. Lean body mass was derived by assuming total body water to be 73.2% of lean body mass, as had been validated in the laboratory. The adipose tissue mass was then obtained by subtraction of the estimated lean body mass from the total weight. Percent body fat was calculated as the adipose tissue mass/body weight. For the OGTT, the beverage containing 0.33 g/ml of glucose was offered to the monkeys shortly before the end of the 16-hour fast. The volume of the beverage offered was based on the monkeys' weight, and the test was considered successful only if the monkeys drank at least 90% of the beverage within a 20-minute period, to give a total glucose dose of approximately 2 g glucose/kg body weight. The actual amount consumed was calculated. Two hours after withdrawal of the beverage, blood was drawn to determine 2 hour levels of both insulin and glucose. To calculate insulin-mediated peripheral glucose uptake, or M-rates, the euglycemic, hyperinsulinemic clamp technique was used as described by Bodkin and colleagues (14). A priming dose of insulin followed by continuous infusion (40 mU x M2 body surface area x min) was administered for 90–120 minutes, with a variable 20% glucose infusion, until steady-state glucose levels were achieved. Steady-state conditions were considered attained after obtaining at least six plasma glucose levels between 80 and 90 mg/dl. The M-rates, corrected for fat-free mass, were calculated from the steady-state plasma glucose levels during the last 30 minutes of the clamp.
Statistical Analysis
Although each monkey in the colony is ideally assessed every 6 months, some monkeys had more, whereas others had fewer, data points. To avoid over- or underrepresentation of an individual within a 2-year age interval, the average value for every variable was calculated over 2-year age ranges. As an example, if a monkey had five determinations of FPG when it was aged between 12.1 and 13.9 years, the average of the five values would be taken to represent its FPG at age range 14. Because there were fewer monkeys younger than age 10 years, values obtained between ages 4.78 and 9.99 years were averaged to constitute the 
10 category, while those obtained after 30 years of age were not included. In effect, the data set consisted of values of the different parameters obtained from monkeys between 4.78 and 29.45 years of age, with a mean observation period per monkey of 7.6 years. The two groups, N and D, were compared over time by performing an analysis of variance (ANOVA) for repeated measures, followed by the Tukey-Kramer multiple comparison test, using the Number Cruncher Statistical Systems software (NCSS 2000; Dr. Jerry L. Hintze, Kaysville, UT). In this analysis, the metabolic category (N vs D) was considered as factor 1, the individual monkey as factor 2 (nested within factor 1), and age range was designated as factor 3. The results of the ANOVA followed by multiple comparison testing revealed (a) whether the mean value for each category, N and D, were significantly different from each other, when the mean is calculated as the average value of a particular parameter, if all the 2-year means from all of the age groups were lumped together for that category; (b) if within the same metabolic category, N or D, the means of the different age ranges were significantly different from one another; (c) if the means of the diabetic group differed significantly from the nondiabetic group of the same age range. For some variables, correlation analysis was performed using the same program. All values are expressed as mean ± the standard error (SE) of the mean.
|
RESULTS
|
|---|
Sixty-five diabetic monkeys and 88 nondiabetic monkeys were included in the study. The diabetic monkeys ranged in age from 5.52 to 29.48 years. The nondiabetic monkeys were aged from 4.78 to 29.45 years.
At ages younger than 10 years, the two groups did not differ significantly in weight, with a mean weight of 11.61 kg ± 0.28 among the N, compared to 12.76 kg ± 0.45 among the D. With advancing age, both the control and diabetic groups tended to gain weight, with peak weights attained at 14–18 years of age. Weights tended to decline thereafter, with the loss of weight being much more pronounced among the diabetic animals than the normal subjects (Figure 1A). The same trend can be seen with the percent body fat, which did not differ between the diabetic subjects and their age-matched controls except at age group 12. With body fat, levels tended to plateau after the initial increase during maturation from early adulthood to middle age in both groups (Figure 1B).
With increase in age, FPG among the nondiabetic subjects remained practically unchanged (Figure 2A). In contrast, among the diabetic monkeys, the fasting plasma glucose levels started to increase with age, becoming significantly higher than the age-matched nondiabetics by age 14 (136.2 mg/dl ± 7.9 vs 64.3 mg/dl ± 7.1; p <.001). Since weight increased in both groups, this implied that the change in glucose status among the diabetic subjects was not likely to be associated with the increase in weight.
In the case of fasting plasma insulin (IRI), mean insulin values in the diabetic group (213.37 µU/ml ± 22.74) were already significantly above those of the nondiabetics (58.57 µU/ml ± 14.51) by age 10 (p <.01). The highest values of IRI were observed among the diabetic monkeys in the 12–14-year age group, when the monkeys were likely to be prediabetic. With further progression of aging and disease, mean insulin values among the diabetic subjects started to wane, eventually becoming less than those of the nondiabetic subjects. By age 30, the IRI among the D was 41.7 µU/ml ± 33.7, compared to 87.6 µU/ml ± 37.7 among the N. Among the N, there did not appear to be any significant age-related change in IRI with aging (Figure 2B). Similar to the results for FPG, HbA1C levels remained practically unchanged with advancing age among the N, but among the diabetic monkeys, levels were already significantly elevated above the age-matched N group by age 16, and continued to be elevated throughout the aging process (Figure 3).
Values of plasma glucose obtained 2 hours after an OGTT-G also remained constant with age among the N (Figure 4A). However, among the D, OGTT-G levels increased after early adulthood, and as early as age 14, were already significantly higher than the levels in age-matched N (193.1 mg/dl ± 14.64 vs 80.6 mg/dl ± 11.14; p <.03). The plasma insulin values obtained at 2 hours during glucose tolerance testing, OGTT-I, also did not vary with age among the N, whereas among the D, insulin values were significantly higher than N as early an age as group 10 as seen in Figure 4B (D: 1373.0 µU/ml ± 156.7 vs N: 221.62 ± 92.0 µU/ml, p <.04).
Consistent with the fasting hyperinsulinemia and the elevated OGTT-I by age 10, diabetic monkeys also demonstrated an exaggerated acute insulin response (AIR) before 10 years of age; thereafter, the response started to wane, so that among the older age group, the response was blunted compared to values obtained in the same group at earlier ages. In contrast, among N, AIR levels were the same between different age groups (Figure 5A).
KG, or the glucose disposal rate following an intravenous glucose tolerance test (IVGTT), appears to decline steeply with age (Figure 5B) if data from the entire population, N and D, were plotted against age. However, closer inspection of KG reveals that among the N, KG remains almost the same after age 14. In contrast, as early as the age 12 group, the KG of the diabetic monkeys were already significantly lower than those of the normal group (D: 2.56 mg/dl/min ± 0.26 vs N: 3.20 mg/dl/min ± 0.13; p .01) and progressively decreases with aging among the diabetic subjects. This suggests that the observed progressive decline in KG with age observed in many previous studies may be due to the contribution of prediabetic subjects who had not yet converted and were therefore misclassified as being "normal." Similarly, the glucose uptake rate during a hyperinsulinemic, euglycemic clamp procedure, or insulin sensitivity index (M-rate), also appears to be fairly unaltered across the age groups among the nondiabetic elderly subjects (Figure 6). Sparseness of data among the diabetic category across age groups prevents us from evaluating age changes among the diabetics. Taken together, the data from both the IVGTT and the euglycemic, hyperinsulinemic clamp suggest that, among those who have aged successfully, the decline in glucose tolerance and insulin sensitivity occurs only during the maturation process from youth to middle age, after which both glucose tolerance and insulin sensitivity remain unchanged.
View larger version (26K):
[in this window]
[in a new window]
|
Figure 6. Changes in mean M-rate (insulin sensitivity) with age in Nondiabetic monkeys. The M-rate among Nondiabetic monkeys showed no significant change after maturation
|
|
From age 18 years, diabetics monkeys also have significantly higher levels of TG compared with nondiabetic subjects (D: 539.42 mg/dl ± 42.6 vs N: 131.4 mg/dl ± 36.2), with values remaining elevated thereafter (Figure 7A). In contrast, mean TG levels did not appear to be associated with age among the control group. Similarly, an age-related increase in total cholesterol was not found among N (Figure 7B), while the D had significantly higher values than N from age 18 (N: 131.8 mg/dl ± 11.7 vs D: 226.0 mg/dl ± 13.2; p <.01).
In both N and D, HDL cholesterol tended to decline with age (Figure 8A). However, after age group 16, values tended to plateau among the nondiabetic monkeys, whereas values continued to decrease among the diabetic subjects. At no age group were the mean HDL values of the diabetic group significantly different from those of age-matched controls. Mean levels of low density lipoprotein (LDL) cholesterol among the N did not change significantly with age (Figure 8B), but among the D, mean levels of LDL cholesterol at age group 28 (122.25 mg/dl ± 12.9) were significantly higher than values obtained from diabetics at age 10 (46.0 mg/dl ± 11.56, p <.05), again suggesting that the age-related increases in LDL cholesterol occur primarily in the diabetic group.
Between age groups 14–26, mean levels of very low density lipoprotein (VLDL) TG among diabetic monkeys were significantly higher than the age-matched nondiabetic group (Figure 9A). Again, no age-related changes could be observed among monkeys in group N, whereas within group D, mean levels of VLDL TG between age groups 14–24 years were significantly higher than levels observed in the same group at age 10.
Very similar to the trend in VLDL cholesterol, VLDL TG (Figure 9B) among the diabetic monkeys were significantly higher than the levels observed among the control monkeys between age groups 14 and 26. Among nondiabetic monkeys, again, there was no association between age and levels of VLDL cholesterol. Within the diabetic group, VLDL cholesterol levels between age groups 14–26 years were significantly higher than the mean level at age group 10.
|
DISCUSSION
|
|---|
It is often assumed that, as one ages, glucose homeostasis is progressively impaired. Studies in the past supported this assumption (3,15). However, Andres posed the question of whether the impairment in glucose homeostasis seen with aging is in fact a reflection of the increased prevalence of diabetes in elderly persons, rather than nonpathologic changes intrinsic to the aging process (1), precisely the question that we have addressed here. The current study shows that, among our nondiabetic subjects, fasting plasma glucose remained practically unchanged with age. Furthermore, glucose levels obtained 2 hours post-OGTT were also the same across the various age groups studied, from groups 10–30. Consistent with these findings, mean levels of glycated hemoglobin or HbA1C did not change significantly with progression of aging. These findings contradict the common perception that a decline in glucose tolerance is characteristic of aging, as suggested by an older review by Davidson (2). In contrast to the results found in the monkeys that had aged successfully, the diabetic monkeys exhibited the rise in fasting plasma glucose, 2-hours post-OGTT glucose, and HbA1C levels that were commonly associated with aging. As early as age group 14, mean values for FPG and OGTT-G, and age group 16 for HbA1C, had already attained levels significantly higher than the nondiabetic controls. Since body weight and body fat increased similarly in both groups during this period, the change in glucose status among the diabetic subjects was not likely to be due to adiposity or weight gain. Other studies support our findings and reported neither difference in glucose nor insulin concentration between young and older subjects following an oral glucose challenge (16,17). Andres and his group (5), analyzing data from the Baltimore Longitudinal Study on Aging, also found little variation in both the FPG and the 2-hour OGTT-G between young (24–39 years) and old normal (60–87 years) participants, but found that OGTT-G was higher among those he called the "old impaired group" who were aged 61–90 years.
Further confirming our finding of preserved glucose homeostasis among our nondiabetic probands, we also found that the glucose disposal rate during an intravenous glucose tolerance test, or KG, changed only minimally between ages 10–30 years, being most conspicuous between early adulthood and monkey middle age. This finding is in agreement with the finding of Reaven and his group (4) in nonobese, physically active, and disease-free human participants, where the decline in glucose tolerance between two groups, young (mean age 33 y) and old (mean age 64 y) was relatively "modest," averaging only 2 mg/dl per decade. Among the diabetic monkeys, however, we saw a steep decline in KG in the same period, and glucose intolerance was already significant by age group 12. Combining data from both diabetic and nondiabetic subjects, as is often the case in human studies, generates a curve that incorrectly suggests a marked deterioration of glucose tolerance with aging.
To determine whether there is a significant decrease in insulin sensitivity with age, we examined the results from the euglycemic, hyperinsulinemic clamp and found that, among normal monkeys, the glucose uptake rate (M-rate) does not change with age per se after middle age if the subjects have neither diabetes nor the metabolic syndrome. This is in agreement with human studies done by Andres and his associates (16), where "differences in (insulin) sensitivity with age" in three groups of participants with mean ages of 26, 52, and 72 years were not observed following a euglycemic clamp (16). Others studies from the same group (5) also showed that the M-rate did not differ statistically among three age groups (24–29 y; 40–59 y; 60–90 y), provided the participants had normal oral glucose tolerance values. A decrease in M was, however, found among the old participants with impaired OGTT-G. The authors suggested that neither total suppression of hepatic glucose production by insulin, nor insulin secretion in response to intravenous hyperglycemia, nor glucose utilization (M), were altered with aging, after adjustment for various other factors that influence insulin sensitivity other than age. Again, the studies by Pacini and colleagues (18–20) demonstrated no changes in insulin sensitivity, FPG, fasting plasma insulin, glucose disappearance rate at basal insulin, and glucose tolerance following OGTT and IVGTT among healthy, nonobese elderly persons compared with young controls, which is similar to the results in our nondiabetic monkeys. These findings suggest that age per se does not contribute to glucose intolerance if other variables are well controlled for to ensure that diabetes is not the cause of the decline.
Comparing our findings with those done in nonhuman primates, Kemnitz and his associates, comparing ad libitum-fed (control) to dietary-restricted rhesus monkeys in captivity (21,22), found that monkeys in the control group at age 17.9 years experienced a decrease in KG as well as insulin-dependent and insulin-independent glucose uptake compared to their levels at 9.6 years. In our study, we also observed a slow decline in both KG and M-rates at these age ranges, which corresponds to a fall from early adulthood to middle age. In their study, the fall in KG and insulin sensitivity was not found to be statistically associated with adiposity in the control monkeys. In our group of nondiabetic monkeys, KG was weakly associated with weight (r = –0.28, p <.001) but not with body fat (r = –0.12, not significant). However, the insulin sensitivity index, M, was significantly correlated with both weight (r = –0.49, p <.001) and body fat (r = –0.50, p <.001) among the control monkeys. In monkeys older than 8 years, there is very little change in lean mass, and thus changes in weight are indicative primarily of increases (or decreases) in fat mass, as shown in Figure 1. Our data therefore suggest that the initial decline in KG and M-rate among normal aging monkeys may be a consequence of the increase in weight and body fat associated with maturation, after which the both KG and M remain virtually unchanged after middle age is attained. This conclusion appears reasonable in the light of a recent Canadian study (23) comparing young (18–35 y) and middle-aged (50–70 y) participants, which proposed that the purported decrease in glucose tolerance and insulin sensitivity with aging correlates more with visceral obesity rather than aging per se.
With regard to baseline and glucose-stimulated insulin secretion, our results show that the mean levels of fasting insulin, levels at 2-hours post OGTT, and the AIR to intravenous glucose challenge do not change with progression of aging among the monkeys who have aged successfully. In humans, it appears that the age effect on insulin secretion is small, and the insulin response to a hyperglycemic clamp is the same in young and old subjects with comparable glucose tolerance (24). In contrast is the remarkable hyperinsulinemia among the diabetic monkeys that had significantly higher mean fasting insulin, OGTT-I, and AIR values than their age-matched controls by age 10. All three insulin variables eventually dwindle with progression of age and diabetes in the diabetic group, while remaining invariable beyond middle age among the nondiabetic group. Since the elevation in insulin secretion precedes fasting and postchallenge hyperglycemia, the decline in KG, the decrease in M-rate, the gain in body weight, and the increase in body fat by at least 2 years, it is unlikely that the hypersecretion of insulin in the diabetic cohort is a manifestation of insulin resistance or a consequence of adiposity. We believe that hyperinsulinemia is one of the earliest prognosticators of animals who will progress into diabetes. This conclusion is supported by a prospective study among Pima Indians with normal glucose tolerance, where fasting hyperinsulinemia was found to be an independent risk factor for the development of diabetes, independent of insulin resistance (25). In the cited study, the diabetogenic actions of hyperinsulinemia also did not appear to be mediated by an increase in body weight, leading the authors to postulate that the hyperinsulinemia may be due to increased vagal stimulation to the pancreas.
Variables associated with lipid metabolism, such as circulating TGs, total cholesterol, and all the lipoprotein fractions (except for HDL cholesterol) remained unchanged with progression of age among our normal monkeys. In the diabetic cohort, dyslipidemia was highly evident, as shown by the elevated levels of TGs, total cholesterol, LDL cholesterol, VLDL cholesterol, and VLDL TG compared to age-matched control animals. While it has been proposed that the elevation in TG levels may be due to augmentation of hepatic TG synthesis secondary to hyperinsulinemia (26), this cannot explain the persistence of dyslipidemia in the later phases of the disease when both insulin level and body weight decline.
Summary
We have shown that among nondiabetic subjects, which neither demonstrated hyperglycemia nor the characteristics of the metabolic syndrome at any period during the study, the decline in glucose tolerance, as reflected by a decline in KG, occurs only in early adulthood up to age 14 years and suffers no further deterioration from middle age to the advanced years. Similarly, insulin sensitivity (M-rate) also declines only from early adulthood to middle age and shows no further reduction with advancing age. Body weight showed a steady increase from youth to middle age, coinciding with the decrease in glucose tolerance and M-rates, which is strongly similar to the human paradigm. Common indicators of glucose and lipid homeostasis, such as FPG, HbA1C, IRI, 2-hour OGTT-G, OGTT-I, AIR, total TGs, total cholesterol, LDL cholesterol, and VLDL cholesterol, did not demonstrate age-related changes in the nondiabetic group. In diabetic monkeys, the expected changes that were indicative of deterioration of glucose and lipid homeostatic function, and which had been previously attributed to aging, were all observed. Discrepancies between other studies using human and nonhuman primate data and our current findings may be due to inherent differences in our comparison group, which consisted of "successful" elderly subjects, having excluded subjects with any history of hyperglycemia or the metabolic syndrome. Also noteworthy is the absence of confounding in our data due to ethnicity, lifestyle, drug history, and dietary composition. Studies that have attempted to control for these confounders have findings that are highly consistent with ours. It has been shown, therefore, that in adult human primates, deterioration of carbohydrate and lipid metabolism are more reflective of the development of disease, rather than of aging.
|
Acknowledgments
|
|---|
The authors are especially grateful to all the Obesity and Diabetes Research Center staff members who contributed to gathering and entering data for the database, including Theresa Alexander, Holly Jermyn, Michelle Izuka, Wallace Evans, and Karen Brocklehurst.
This work was supported by contract NO1-AG-02100 from the National Institutes of Health National Institute on Aging to the Obesity and Diabetes Research Center at the University of Maryland at Baltimore, School of Medicine; Dr. Barbara Caleen Hansen, Principal Investigator.
|
Footnotes
|
|---|
Decision Editor: James R. Smith, PhD
Received March 17, 2004
Accepted July 15, 2004
|
References
|
|---|
- Andres R. Aging and diabetes. Med Clin North Am. 1971:;55:835-846.[Medline]
- Davidson MB. The effect of aging on carbohydrate metabolism: a review of the English literature and a practical approach to the diagnosis of diabetes mellitus in the elderly. Metabolism. 1979:;28:688-705.[Medline]
- DeFronzo RA. Glucose intolerance and aging. Diabetes Care. 1981:;4:493-501.[Abstract]
- Reaven GM, Chen N, Hollenbeck C, Chen YD. Effect of age on glucose tolerance and glucose uptake in healthy individuals. J Am Geriatr Soc. 1989:;37:735-740.[Medline]
- Elahi D, Muller DC, McAloon-Dyke M, et al. The effect of age on insulin response and glucose utilization during four hyperglycemic plateaus. Exp Gerontol. 1993:;28:393-409.[Medline]
- Muller DC, Elahi D, Tobin JD, Andres R. The effect of age on insulin resistance and secretion: a review. Semin Nephrol. 1996:;16:289-298.[Medline]
- Elahi D, Muller DC, Egan JM, et al. Glucose tolerance, glucose utilization and insulin secretion in ageing. Novartis Found Symp. 2002:;242:222-242 discussion 42–46.[Medline]
- Rowe JW, Kahn RL. Human aging: usual and successful. Science. 1987:;237:143-149.[Abstract/Free Full Text]
- Hansen BC. Obesity and diabetes: the natural history. In: Turtle JR, Kaneko T, Osato S, eds. Diabetes in the New Millennium. Sydney: Endocrinology and Diabetes Research Foundation of University of Sydney; 1999:151–160.
- Standards of medical care in diabetes. Diabetes Care. 2004:27(Suppl 1):S15–S35.
- Reinauer H, Home P, Kanagasabapthy A, Heuck C. Laboratory Diagnosis and Monitoring of Diabetes Mellitus. Geneva: World Health Organization; 2002.
- Hansen BC, Bodkin NL. Heterogeneity of insulin responses: phases leading to type 2 (non-insulin-dependent) diabetes mellitus in the rhesus monkey. Diabetologia. 1986:;29:713-719.[Medline]
- Bodkin NL, Hannah JS, Ortmeyer HK, Hansen BC. Central obesity in rhesus monkeys: association with hyperinsulinemia, insulin resistance, and hypertriglyceridemia? Int J Obesity. 1993:;17:53-61.
- Bodkin NL, Metzger BL, Hansen BC. Hepatic glucose production and insulin sensitivity preceding diabetes in monkeys. Am J Physiol. 1989:;256:(Endocrinol Metab): E676-E681.[Medline]
- DeFronzo RA. Glucose intolerance and aging: Evidence for tissue insensitivity to insulin. Diabetes. 1979:;28:1095-1101.[Medline]
- Andres R, Tobin JD. Aging and the disposition of glucose. Adv Exp Med Biol. 1975:;61:239-249.[Medline]
- Duckworth WC, Kitabchi AE. Direct measurement of plasma proinsulin in normal and diabetic subjects. Am J Med. 1972:;53:418-427.[Medline]
- Pacini G, Valerio A, Beccaro F, et al. Insulin sensitivity and beta-cell responsivity are not decreased in elderly subjects with normal OGTT. J Am Geriatr Soc. 1988:;36:317-323.[Medline]
- Beccaro F, Pacini G, Valerio A, et al. Age and glucose tolerance in healthy subjects. Aging (Milano). 1990:;2:277-282.[Medline]
- Viviani GL, Pacini G. Reduced glucose effectiveness as a feature of glucose intolerance: evidence in elderly type-2 diabetic subjects. Aging (Milano). 1999:;11:169-175.[Medline]
- Ramsey JJ, Colman RJ, Binkley NC, et al. Dietary restriction and aging in rhesus monkeys: the University of Wisconsin study. Exp Gerontol. 2000:;35:1131-1149.[Medline]
- Gresl TA, Colman RJ, Roecker EB, et al. Dietary restriction and glucose regulation in aging rhesus monkeys: a follow-up report at 8.5 yr. Am J Physiol Endocrinol Metab. 2001:;281:E757-E765.[Abstract/Free Full Text]
- Imbeault P, Prins JB, Stolic M, et al. Aging per se does not influence glucose homeostasis: in vivo and in vitro evidence. Diabetes Care. 2003:;26:480-484.[Abstract/Free Full Text]
- Chang AM, Halter JB. Aging and insulin secretion. Am J Physiol Endocrinol Metab. 2003:;284:E7-E12.[Abstract/Free Full Text]
- Weyer C, Hanson RL, Tataranni PA, et al. A high fasting plasma insulin concentration predicts type 2 diabetes independent of insulin resistance: evidence for a pathogenic role of relative hyperinsulinemia. Diabetes. 2000:;49:2094-2101.[Abstract/Free Full Text]
- Kemnitz JW, Goy RW, Flitsch TJ, et al. Obesity in male and female rhesus monkeys: fat distribution, glucoregulation, and serum androgen levels. J Clin Endocrinol Metab. 1989:;69:287-293.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
M.-A. Cornier, D. Dabelea, T. L. Hernandez, R. C. Lindstrom, A. J. Steig, N. R. Stob, R. E. Van Pelt, H. Wang, and R. H. Eckel
The Metabolic Syndrome
Endocr. Rev.,
December 1, 2008;
29(7):
777 - 822.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. O. Chavez, J. C. Lopez-Alvarenga, M. E. Tejero, C. Triplitt, R. A. Bastarrachea, A. Sriwijitkamol, P. Tantiwong, V. S. Voruganti, N. Musi, A. G. Comuzzie, et al.
Physiological and Molecular Determinants of Insulin Action in the Baboon
Diabetes,
April 1, 2008;
57(4):
899 - 908.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Methods
) and Diabetic (D) (•) monkeys. (A) Weights increase in both N and D from early to middle age, then declines among the late elderly, with a greater decrease seen among D. (B) Body fat increases with maturation in both groups, N and D, and then plateaus from middle age. *Significantly different from the mean value at age group 10, among N; 
Significantly different from the mean value of age-matched N
